Application of Reinforcement Learning in Cognitive Radio Networks: Models and Algorithms

Table 8

RL model for the channel auction scheme [16].

State	𝑠^𝑖𝑡∈𝑆={0,1,…,𝐿𝑏⋅𝐿𝑐}, each state represents a two-tuple information composed of buffer fullness index 𝐿𝑏 and credit ratio index 𝐿𝑐
Action	𝑎^𝑖𝑡∈𝐴={1,2,…,𝐿𝑎}; each action represents the amount of a bid for white spaces
Reward	𝑟^𝑖𝑡+1𝑠^𝑖𝑡+1=Positivevalueoftheamountofdatasent,ifsuccessfulbidNegativevalueoftheamountofdatathatcouldhavesent,ifunsuccessfulbid